OBJECTIVE: Our study was to assess the validity of SAPS II (New Simplified Acute Physiology Score) to predict the probability of in hospital mortality, in a cohort of patient admitted to a medical intensive care unit. STUDY DESIGN: Prospective study. PATIENTS AND METHODS: Out of 467 the 525 patients admitted were included. SAPS score and in hospital mortality prediction were calculated for each of them. RESULTS: In this group, SAPS II offered a satisfactory discrimination power with an area under the curve of 0.843. However, calibration showed a lack of fit (chi 2 = 28.5, P < 0.001), with an overall under prediction of mortality (observed versus expected ratio of 1.12). CONCLUSION: This SAPS II lower predicting accuracy in a specific population and for individual outcome prediction may reduce its interest in clinical decision-making.
OBJECTIVE: Our study was to assess the validity of SAPS II (New Simplified Acute Physiology Score) to predict the probability of in hospital mortality, in a cohort of patient admitted to a medical intensive care unit. STUDY DESIGN: Prospective study. PATIENTS AND METHODS: Out of 467 the 525 patients admitted were included. SAPS score and in hospital mortality prediction were calculated for each of them. RESULTS: In this group, SAPS II offered a satisfactory discrimination power with an area under the curve of 0.843. However, calibration showed a lack of fit (chi 2 = 28.5, P < 0.001), with an overall under prediction of mortality (observed versus expected ratio of 1.12). CONCLUSION: This SAPS II lower predicting accuracy in a specific population and for individual outcome prediction may reduce its interest in clinical decision-making.
Authors: André Carlos Kajdacsy-Balla Amaral; Fábio Moreira Andrade; Rui Moreno; Antonio Artigas; Francis Cantraine; Jean-Louis Vincent Journal: Intensive Care Med Date: 2005-01-25 Impact factor: 17.440